Metraipolis › archive › dc_d48aada7d3ea
PersonaCUABench: Benchmarking Computer-Use Agents Serving Diverse Personas — arXiv:2610.22990
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.22990
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T08:43:05.797Z
- SHA-256 of the archived text
e9a41371da6cf09991826862df3f00a10c8d26ea76c2c2a5b6d47d3bd2ecf762- Stable URL of this page
https://metraipolis.com/documento/dc_d48aada7d3ea
How to cite
PersonaCUABench: Benchmarking Computer-Use Agents Serving Diverse Personas — arXiv:2610.22990 — arXiv (en), https://arxiv.org/abs/2610.22990 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d48aada7d3ea (Metraipolis archive dc_d48aada7d3ea, sha256 e9a41371da6cf09991826862df3f00a10c8d26ea76c2c2a5b6d47d3bd2ecf762)
BibTeX
@misc{dcd48aada7d3ea,
title = {PersonaCUABench: Benchmarking Computer-Use Agents Serving Diverse Personas — arXiv:2610.22990},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.22990},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d48aada7d3ea; sha256 e9a41371da6cf09991826862df3f00a10c8d26ea76c2c2a5b6d47d3bd2ecf762},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.22990. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Computer-use agents (CUAs) are commonly evaluated against fixed task specifications, but users may omit requirements, add constraints, or require approval before consequential actions. Completing the benchmark instruction does not guarantee serving the person who asked. We introduce PersonaCUABench, which holds 74 tasks across 11 domains fixed and varies only who asks. A persona agent, playing one of 44 personas in four groups, rewrites each instruction in that user's own words and answers the CUA's questions, while the CUA sees only the dialogue, never the profile or the original instruction. Across nine backbones (3,330 runs), mean task completion falls by 36% relative to the original requests. The losses have two sources: in 59.9% of cells that lose points, the agent engages with what the user asked for and still gets it wrong, and 11.2% are stops the user requested that the checker scores as failures. Even when a persona request is fully completed, the user gives the top satisfaction rating in only 62.7% of cells, against 84.1% for the original requests. CUAs should therefore be evaluated not only on task completion but also on whether they meet the requirements and constraints users express. Code: this https URL . Project page: this https URL Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Human-Computer Interaction (cs.HC) ; Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2610.22990 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-19T08:39:10.138Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).